Archive for the ‘Misc’ Category

After a long time…

August 31, 2009

It’s been a long time since I’ve posted here and it’s all my fault. A lot happened but the markets now are stabilizing, jobs are safer and we can go back to writing blogs and sipping coffee … I suppose.
I saw recently one comment to a previous problem I posted here and I’m trying to contact the poster to provide more explanation, as I’m intrigued. I didn’t have to think more about the problem and solutions, but if he comes back to me with a detailed answer, I will try to digest it and publish it.
I will also try to write more here and resurrect this blog…


Investment banking shocker

September 16, 2008

Over the weekend, Lehman and Merill are both gone, one bust, the other bought by Bank of America. That’s about 5000 people without jobs in the City of London. I’ve heard of desperate calls for jobs from techies in Lehman. Some of them will find jobs with other investment banks, but I suspect a lot of them will have to quit banking and work for other technology companies. It’s not the end of the world of course…

A better review of Chrome

September 5, 2008

Testing Chrome

September 3, 2008

I have installed Google Chrome today and played with it for about 5 minutes. First impression was … ok, it looks like a minimal browser, no menus, no fluff, a sort of F11-ed browser. Unfortunately, Chrome wasn’t able to render even some of Google’s own websites. It couldn’t render properly the Google finance website. And then, from the main site, clicking on other Google Apps in their bar, would take me to a blank window each time. Loading brought me a site without all the images – particularly the main headline photo wasn’t being displayed.

So I thought enough was enough. It’s a long way ahead for Chrome at the moment, especially that Firefox 3 set the bar so high. Back to Firefox for me…

And the first screenshots…

September 2, 2008

September 2, 2008

A very “Cool a la Google” presentation of their Chrome browser.

Very interesting stuff and exciting times ahead.

Why do we need another browser?

September 2, 2008

So Google thinks there are too few browsers out there. Ask any web designer and they’ll tell you: There are too many. Testing applications on IE and Firefox is cumbersome enough as it is. Maybe that’s about to change with IE8 as sources close to Microsoft said IE8 is more compliant. The problem will not go away so easily though, as more and more people are reluctant to upgrade their MS components (see the Vista flop).

So why does Google feel that it needs its own browser? Especially that it is Firefox’ main sugar daddy.

Perhaps because the browser will be the battlefield for the next paradigm shift. Since Web 2.0 popped its over-hyped head, people started dreaming again about the Internet and started dreaming about a world where applications would run in the browser, making the underlying OS more or less irrelevant. It’s one major weakness for Microsoft. Microsoft has a near monopoly on the desktop, perhaps in the same way that once Apple had a monopoly of graphical user interfaces and IBM of personal computers. They all lost their monopolies by being complacent and not driving the market. A market leader has the responsibility of driving the market, not just milk the customers.
Microsoft managed to displace Apple, IBM and many others by being quick to market (with half baked incremental solutions) and by vapourware (announcing products way before they’re ready in order to kill off demand for competitor software). But Microsoft is now perhaps in a position whereby the only direction is downwards.
Microsoft for too long has ignored the browser market, instead of cultivating it and driving it forward as a true market leader. And that is turning against them now.

Google no doubt is very aware of this vulnerability and the new browser is their show of commitment to take over where Microsoft seemed to have left it off. Google would like to drive forward the browser market and in the process deliver a better platform for delivering web apps that can compete better with the MS Cash Cows (Office in particular). There’s a long way, but it all looks very promising. Improvements to Javascript engines made it run very fast indeed, opening the roads for applications that were not possible before. I still remember the days when a for loop in JS over some DOM elements would take 30 seconds or so. Those days are gone and the future for web apps looks great.

The questions I have are:
1. What will happen to Firefox? – I love Firefox
2. What will be Microsoft’s reaction? – will IE8 be any good? will it be any innovative or just a “me too” browser?
3. How fast is Google’s Chrome going to be compared to Firefox and IE8
4. How many Google specific features (e.g. Code gears) will Chrome have compared to the compliant browsers?

Deperation, thy name … Vista

July 29, 2008

I mean Mojave. I mean Vista. I really mean Vista.
As someone already mentioned, the experiment uses Adobe Flash and not Silverlight. Or could you be convinced it’s really Silverlight too?

Unilateral and bilateral searching

July 16, 2008

Imagine you want to buy a new web camera. You go onto google, type ‘web camera’, you click the first link that takes you to website1 and you look at the price. Let’s assume you know exactly what you want, so you only check the price of the model that you want. You face a choice: buy now or keep searching. You choose to write down the price and return to the google result listing. You click on the second link that takes you to website2 and you repeat the process. You compare the lowest price of the two and again you have a choice: Buy the best price or keep searching.

In theory you could repeat the process ad infinitum, or until you exhaust the whole google/yahoo/msn etc. databases and you go through all the providers. Then you can continue the search by visiting the high street shops in your city and compare the prices in the same fashion. In the end you will find your best buy.

The main problem with the process above is that it is time consuming. That means costly. You can estimate the cost of your time spent on this by putting a value on it. For instance, if you earn X$ an hour, you can easily calculate in the end how much the search has cost you. At any point, you can calculate an average cost per new search by looking at how costly previous searches were. You might discover that the longer you search, the more expensive each search becomes as a lot of the information from the search engines contains a lot of false positives, the further you navigate away from the first page.

So, at any point, you can calculate the cost of you choice between buying now and the cost of searching and even a probability of finding a better price. You can calculate the point when you can stop the search in terms of the cost of searching and the cost of your time. As such, you can stop when the cost of the next search is greater than the cost of your time spent on the search. You also have to consider the probability that the new search will improve the best price you already discovered.

That’s an unilateral search and it’s characterized by a dynamic side, You, searching for the web camera, which web camera is the static side, which doesn’t change (You can always return to a previous result).

Bilateral searching is different from unilateral searching in that both sides are dynamic and searching.

Imagine you’re looking to buy a house. You can repeat the process described in the unilateral search, but as you keep searching, the other party keeps searching too. So in the end, it might not always be possible to return to your previous best buy. Your previous best buy might decide to sell the house to someone else who offers a better price, while you keep searching.

As such, the calculation of cost for bilateral searching is more complicated and it needs to take into consideration the risk of losing the best buy while you keep searching for price improvement. That means usually that you would stop sooner in a bilateral search than in a unilateral search. Alternatively, you might not lose best buys or take much longer to trade as you lose some opportunities and you have to restart your search.

This risk of losing the current offer also forces us to settle for good buys rather than best buys.

On the house market, this is evident as buyers are quick to put offers in a sellers’ market than in a buyers’ market. In a buyers’ market, they take more time to search, as they know the risk of the ‘best buy’ going away is diminished.

The bilateral search applies to many searches in real life: house purchases, renting, buying unique goods (e.g. art, second hand cars, etc.), financial trading of securities, and many others. Interestingly, this applies in a more general context and it affects searching for a partner for example…